A total of 17 papers were identified that evaluated or described wellbeing measures for Indigenous children and youth. Eight (47%) publications came from Australia, five (29%) from Canada, three (18%) from Aotearoa/New Zealand, one (6%) from the U.S.A, and none from Norway. The 17 papers represented nine studies and described the development and/or use of 14 measurement tools, employed across 27 applications. The publications are summarised in Table 1.
Table 1
Summary of the publications
Publication (yr)
|
Aim
|
Country
|
Measures
|
Target population of Indigenous children and youth
|
Purpose
|
General population
|
Specific mental health concerns
|
Development of Indigenous-specific tool
|
Trials of Western measures
|
Identification of wellbeing risks
|
Identification of wellbeing improvement
|
Clark 2014
|
✓
|
NZ
|
Children’s Global Assessment Scale (C-GAS); Strengths and Difficulties Questionnaire (SDQ); Substance Abuse Choices Scale (SACS)
|
✓
|
✓
|
|
|
|
✓
|
Coffin 2019
|
✓
|
Aus
|
Strengths and Difficulties Questionnaire (SDQ)
|
✓
|
|
|
✓
|
|
|
Goodyear-Smith 2016
|
✓✓
|
NZ
|
YouthCHAT; Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) for smoking; Generalised Anxiety Disorder 7 (GAD-7); Patient Health Questionnaire 9 (PHQ-9); Substance Abuse Choices Scale (SACS)
|
✓
|
|
✓
|
|
|
|
Good-year-Smith 2017
|
Harriss 2018
|
✓
|
Aus
|
Adapted Patient Health Questionnaire 9 (aPHQ-9)
|
✓
|
|
|
|
|
✓
|
Mitchell 2011
|
✓
|
USA
|
Kessler Distress Scale (K6+)
|
✓
|
|
|
✓
|
✓
|
|
Sabbioni 2018
|
✓
|
Aus
|
Outcome Rating Scale (ORS); Session Rating Scale (SRS)
|
✓
|
✓
|
|
✓
|
|
✓
|
Thomas 2010
|
✓
|
Aus
|
Strong Souls; Westerman Aboriginal Symptoms Checklist- Youth (WASC-Y); Kessler Distress Scale (K6+)
|
✓
|
|
✓✓
|
|
✓
|
|
Williamson 2010
|
✓
|
Aus
|
Strengths and Difficulties Questionnaire (SDQ)
|
✓
|
|
|
✓
|
|
|
Williamson 2014
|
|
Williamson 2016
|
|
✓
|
|
Williamson 2018
|
✓
|
Aus
|
ED presentations and hospital admissions with primary mental health diagnosis
|
|
✓
|
|
|
Young 2011
|
✓
|
Canada
|
Aboriginal Children’s Health and Well-Being Measure (ACHWM)
|
✓
|
|
✓
|
|
|
|
Young 2015
|
Young 2015a
|
Young 2016
|
Young 2016a
|
Type of measure, target population and purpose documented in the included publications
All 17 publications across four countries identified and described available wellbeing measurement tools [50–66]. Fourteen of the 17 publications (82%) targeted the general population of Indigenous children and youth, and three (18%) targeted Indigenous children and youth who had already reported mental health concerns [50, 62, 65].
Eight publications (three studies) described the development of an Indigenous-specific tool. Young, Wabano [56] described the co-creation with Indigenous Canadian children and youth, and other community members of the Aboriginal Children’s Health and Well-Being Measure (ACHWM), and establishment of its validity [57], appropriateness [67] internal consistency and test–retest validity [68], specificity, negative predictive value, and sensitivity [58]. Thomas, Cairney [53] described the development of Strong Souls with Indigenous Australian youth; it was piloted alongside another Indigenous-specific tool, the Westerman Aboriginal Symptoms Checklist- Youth (WASC-Y). Goodyear-Smith described the acceptability and utility of the YouthCHAT, a youth-adapted version of the previously New Zealand-developed electronic Case-finding and Help Assessment Tool (e-CHAT) for primary care populations [51, 66].
Five publications described trials of the appropriateness and utility of standardised Western measures. Coffin [59] used the SDQ to augment observational improvements in the self-regulation, self-awareness, and socialisation skills of Indigenous Australian youth after their participation in an equine assisted learning program. Mitchell and Beals [52] found the Kessler-6 (K-6) to be a useful complement to more traditional clinical decisions of presence or absence of disorders in native American youth. The K-6 results helped clinicians to understand the impact of psychological disorders more comprehensively, and to make more informed treatment recommendations and plans. Sabbioni, Feehan [65] trialled the Outcome Rating Scale (ORS) and Session Rating Scale (SRS) with Indigenous Australian youth and found that they were simple, brief, and appropriate, and successful in capturing clients’ perspectives about their treatment progress and therapeutic bond; hence they represented a valuable alternative approach to standardised measures. Williamson, Redman [54] assessed the acceptability of the Strengths and Difficulties Questionnaire (SDQ). Indigenous Australian parents and Indigenous staff such as research assistants, youth workers, health staff and education officers reported that the SDQ was acceptable for use in Indigenous Australian Community Controlled Health Organisations, although changes were needed to the wording of some questions and the response scale to improve cultural appropriateness and clarity. A further publication from the same study, however, suggested value in limiting reliance on the peer relationships subscale and focusing instead on the SDQ total difficulties score [69].
Two publications from the same study identified wellbeing risks faced by Indigenous Australian children and youth. Williamson, D'Este [55] administered the SDQ to carers of Indigenous Australian youth clients of four ACCHOs and found that 72% were not at high risk for emotional or behavioural problems. After adjusting for demographic and health characteristics, the factors associated with good mental health were having a carer who was not highly psychologically distressed; not suffering from frequent chest, gastrointestinal or skin infections; and eating two or more servings of vegetables per day. Being raised by a foster carer and having lived in four or more homes since birth were associated with significantly lower odds of good mental health. Later, Williamson, Skinner [62] sought to inform policy and programs to reduce rates of mental illness by examining mental health-related emergency department (ED) presentations and hospitalisations of children recruited through four ACCHOs. They linked population health datasets to find that over a median of 6-year follow-up, there were 96 ED presentations affecting 62 children (10.7/1000 person-years) and 49 hospitalisations for mental health conditions affecting 34 children (5.5/1000 person-years). Presentations and admissions increased with age. Risk factors for ED presentation were living in foster care; high baseline child emotional/ behavioural problems; and caregiver chronic health conditions. Hospitalisations significantly increased when caregivers were unemployed and/or had chronic health problems.
Three publications provided evidence of youth wellbeing improvement outcomes. Clark, Johnson [50] evaluated a referral-based free counseling support service for New Zealand youth with mild to moderate mental health problems and reported significant improvements in global social and psychiatric functioning measured by C-GAS; reduced risk of clinically significant mental health concerns measured by SDQ (p < .001); and reduced use and impact of drugs/alcohol measured by SACS (p < .001). Harriss, Kyle [60] trialled a clinician-administered adapted PHQ-9 measure of depression as part of the Yarrabah Young Person’s Health Check in 2016 and found that one-in-five young people had moderate–severe symptoms or self-harm ideation in the previous 2 weeks. They were referred to the mental health service and followed up by trained staff. The PHQ-9 screening process provided valuable ‘space’ to facilitate communication on sensitive issues and was straightforward and well accepted by staff and youth. Sabbioni, Feehan [65] documented the effectiveness of a culturally sensitive model within YouthLink, a state-wide mental health service program for young people in Western Australia. They found pre-post improvements in youth-administered ORS scores across the Western Australian YouthLink treatment period with a treatment effect size of 0.69. Overall, 65% of clients improved: 35% achieved clinical recovery and 30% achieved clinical cut-off by the final session, but 25% showed no change and 10% deteriorated. Therapeutic alliance (SRS scores) also increased between the first and last session and the final SRS correlated significantly with final ORS suggesting an association between therapeutic alliance and treatment outcome, although the difference lay just outside of statistical significance.
Fourteen measurement instruments were documented in the included publications. Their characteristics are outlined in Table 2.
Table 2
Characteristics of measurement instruments
Measurement instrument
|
Country
|
Relational strength-based constructs of Indigenous wellbeing
|
Developed for Indigenous children/youth
|
Administered through child/ youth self-report
|
Valid and reliable
|
|
Aboriginal Children’s Health and Well-Being Measure (ACHWM)
|
Canada
|
√
|
√
|
√
|
√
|
|
Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) for smoking
|
NZ
|
X
|
X
|
X
|
X
|
|
Children’s Global Assessment Scale (C-GAS)
|
NZ
|
√
|
X
|
X
|
√
|
|
ED presentations and hospital admissions with a primary mental health diagnosis [62]
|
Aus
|
X
|
X
|
X
|
X
|
|
Generalised Anxiety Disorder 7 (GAD-7)
|
NZ
|
X
|
X
|
X
|
√
|
|
Kessler Distress Scale abridged version (K6+)
|
Aus, USA
|
X
|
X
|
X
|
√
|
|
Outcome Rating Scale (ORS)
|
Aus
|
√
|
X
|
√
|
√
|
|
Patient Health Questionnaire 9 (PHQ-9 and a-PHQ-9)1
|
Aus
NZ
|
X
|
X
|
X
|
√
|
|
Session Rating Scale (SRS)
|
Aus
|
√
|
X
|
√
|
√
|
|
Strengths and Difficulties (SDQ)
|
Aus
NZ
|
~
|
X
|
X
|
√
|
|
Strong Souls
|
Aus
|
~
|
√
|
√
|
√
|
|
Substance Abuse Choices Scale (SACS)
|
NZ
|
X
|
X
|
X
|
√
|
|
Westerman Aboriginal Symptoms Checklist- Youth (WASC-Y)
|
Aus
|
X
|
√
|
√
|
√
|
|
YouthCHAT
|
NZ
|
~
|
√
|
√
|
X
|
|
1. a-PHQ 9 was adapted for use with an Australian Indigenous community[70] |
Adherence to the relational strength-based constructs of Indigenous wellbeing
Of the 14 measurement instruments, only four measured solely positive domains of wellbeing. These were the ORS (personal wellbeing, relational wellbeing, social wellbeing, and global wellbeing), SRS (elements of treatment sessions, including the therapeutic relationship, goals, topics, and global rating) [65]; ACHWM (spiritual, emotional, physical and mental health) [56, 57, 68]; and C-GAS (psychosocial functioning at home, at school, and with peers) [50]. Four measurement instruments measured a mixture of strengths and problems, including: Strong Souls (depression, anxiety, suicide risk, drug and alcohol use, impulsivity, anxiety and cultural resilience) [53]; WASC-Y (depression; suicidal behaviours; alcohol and drug usage; impulsivity, hyperactivity and agitation; anxiety; self-esteem; resilience to mental ill health)[53, 71]; SDQ (conduct problems, hyperactivity-inattention, emotional symptoms, peer problems, and prosocial behaviour) [50, 54, 55, 59, 69] and YouthCHAT (substance abuse, depression, anxiety, gambling, sexual orientation, sexual behaviour, exposure to abuse, anger control, physical inactivity, and readiness to change) [51, 66]. The other six measurement instruments focussed on identifying problems, including: the K-6 (feeling sad, nervous, restless or fidgety, hopeless, everything is an effort, or worthless) [52, 53]; SACs (substance use frequency, patterns, abuse and chemical dependency) [50, 51]; GAD (generalised anxiety) [51]; PHQ-9/a-PHQ-9 (depressive symptoms) [60]; ASSIST (smoking); and ED admissions and hospitalisations (hospitalisable mental health diagnosis) [62].
None of the instruments measured all four domains of Indigenous wellbeing as specified by McClintock, King [18]. All but one instrument (the C-GAS) measured some aspects of individual wellbeing (physical, mental and emotional). However, only four instruments encompassed aspects of collective wellbeing such as family, kinship/tribal, and community domains – the C-GAS (psychosocial functioning at home, at school, and with peers), the ORS (relational wellbeing, social wellbeing, and global wellbeing), SRS (the therapeutic relationship, global rating) and SDQ (peer problems, prosocial behaviour)[50, 54, 55, 65, 69]. Only one instrument encompassed an aspect of cultural wellbeing such as language, cultural knowledge and cultural practice domains; Strong Souls included a measure of cultural resilience [53]. Finally, only one measurement instrument encompassed an aspect of spiritual wellbeing such as spirituality, ancestors and land domains; ACHWM included a measure of spiritual health [56–58, 63, 64].
Developed for Indigenous populations
Four of the 14 wellbeing measures had been developed specifically for Indigenous populations. These were the Canadian ACHWM [56–58, 63, 64], Australian Strong Souls [53] and WASC-Y [53], and Aotearoa/New Zealand YouthCHAT [51, 66]. The other 11 were standard Western tools, two of which had been adapted for Australian Indigenous populations [59, 60], and four tools had been translated into an Indigenous (Māori) language [51, 66].
Reporting of outcomes
Eleven of the 18 publications described patient-reported-outcomes (PROs) or child or youth self-report, with one exhibiting both patient-reported and non-patient-reported outcomes. The publications that included patient-reported outcomes focussed on the development of the Indigenous-specific YouthCHAT, Strong Souls, and ACHWM [51, 53, 56–58, 66–68], or the trialling of Western wellbeing measures, the K-6, ORS and SRS, with Indigenous children and youth populations [52, 65], and the identification of actual wellbeing outcomes for Indigenous children and youth [65].
Six publications described non-patient-reported outcomes (non-PRO). Non-patient-reported outcomes included parent/carer or clinician reporting or assessment on behalf of a child/youth. The studies for which non-PRO were used either trialled the Western SDQ measure with Indigenous children and youth populations [54, 69], used the SDQ and ED admissions/ hospitalisations to identify risk [55, 62] or actual wellbeing outcomes for Indigenous children and youth [50, 60].
Reliability and validity
Of the 17 publications, 10 papers [50–55, 65, 66, 68, 69] reported the reliability of 12 measurement instruments (the ACHWM, C-Gas, GAD, K-6, ORS, PHQ9, SDQ, SRS, Strong Souls, SACS, WASC(Y) and YouthCHAT. The most tested type of reliability was internal reliability.
The validity of the measure explored was described in 11 publications [50–54, 56, 57, 65–67, 69]. There was no consistency in the types of validity reported, rendering comparison impossible. The types included content, face, convergent, incremental, criterion, construct, and discriminant validity. Some studies reported the validity of the measure used with mainstream populations, but not for the study population [e.g. 50].